Background: Defensive medicine is characterized by medical decisions made primarily as a precaution against potential malpractice claims. For psychiatrists, professional responsibility encompasses not only the appropriateness of diagnosis and treatment but also the effects of their interventions on patients and their behaviors. Objective: To investigate the socio-demographic, educational, and occupational characteristics and work-related attitudes that may serve as predictors of defensive medicine among Italian psychiatrists. This research extends the results of a previous analysis based on a national survey. Methods: A secondary analysis of the database of a national survey on attitudes and behaviors of Italian psychiatrists regarding defensive medicine and professional liability was performed for this study. Results: Among 254 surveyed psychiatrists, 153 admitted to practicing defensive medicine, while 101 had this attitude with less than half of their patients. The first group was predominantly comprised of women (p = 0.014), who were younger in age (43.34 y 9.89 vs. 48.81 y 11.66, p < 0.001) and had fewer years of professional experience (12.09 y ± 9.8 vs. 17.46 y ± 11.2, p < 0.001). There were no significant differences in prior involvement in complaints (p = 0.876) or the usual place of work (p = 0.818). The most prominent predictors for practicing defensive medicine were (1) considering guidelines and good clinical practices not only for their clinical efficacy but also or exclusively for reducing the risk of legal complaints for professional liability (OR = 3.62; 95%CI, 1.75–7.49), and (2) hospitalizing patients with violent intentions even if not warranted according to their mental state (OR = 2.28; 95%CI, 1.50–3.46, p < 0.001). Prioritizing protection from professional liability over patients’ actual needs in prescribing or adjusting drug dosages and in involuntary hospitalization, as well as prescribing lower dosages than recommended for pregnant patients, were identified as additional predictors. Finally, years of professional experience exhibited a protective function against defensive practices. Conclusions: Psychiatrists advocate the need to implement a ‘risk management culture’ and the provision of more balanced duties in order to ensure ethical and evidence-based care to their patients. A particular source of concern stems from their professional responsibility towards not only the health of patients but also their behavior. However, these aspects conflict with a limited potential for assessment and intervention based on effective clinical tools. A reform of professional liability that considers the specificities of patients cared for by mental health services could contribute to reducing the risk of defensive medicine.